Raw model output on the native 0.25° grid — no MOS, no bias correction, no post-processing. Scores understate operational forecast quality. Read the fairness statement →

CastCheck methodology v0.2 · data through 2026-08-30

Minneapolis-St Paul · GraphCast (IFS init) · lead day 1

KMSP · graphcast_ifs · lead 1

Target day = initialization date + 1 day. This page is a permanent link: the numbers change as history accumulates, the address and the methodology version do not.

Scores use the latest model version segment, GRAP_v100, starting 2026-01-01 — days before that ran a different version of the model and are excluded rather than silently pooled. Of the scored days, 1 carry a QC flag on the observation.

JSON for this page CSV of the daily errors Cite this page

Every number for this combination

All four windows × both initializations × both interpolations × both variables. Errors are computed in °C and shown in °F; CI is a 95 % moving-block bootstrap interval. MAE debiased and skill, debiased remove the constant offset of the window first, so a station that the model always reads a degree too warm is not scored as if it had no skill — compare the two skill columns to see how much of the error is a fixed offset. n QC next to n counts days whose observation carries a QC flag.

VariableInitMethod Windown MAE °F MAE debiased Bias °F RMSE °F ±1 °F±2 °F±3 °F Skill Skill, debiased Period
tmax00Zbilinear30d27 1 QC6.33 [5.38, 7.09]2.04-6.33 [-7.09, -5.38]6.74 [5.81, 7.41]0% [0%, 0%]0%11%-1.34+0.252026-08-02 → 2026-08-30
tmax00Zbilinear90d27 1 QC6.33 [5.24, 7.18]2.04-6.33 [-7.18, -5.24]6.74 [5.61, 7.47]0% [0%, 0%]0%11%-1.34+0.252026-08-02 → 2026-08-30
tmax00Zbilinear365d33 1 QC5.82 [4.49, 6.95]2.16-5.82 [-6.95, -4.49]6.32 [5.02, 7.29]0% [0%, 0%]3%18%-0.90+0.282026-01-02 → 2026-08-30
tmax00Zbilinearall33 1 QC5.82 [4.44, 6.89]2.16-5.82 [-6.89, -4.44]6.32 [4.95, 7.24]0% [0%, 0%]3%18%-0.90+0.282026-01-02 → 2026-08-30
tmax00Znearest30d27 1 QC6.29 [5.31, 7.09]2.08-6.29 [-7.09, -5.31]6.74 [5.78, 7.41]0% [0%, 0%]4%11%-1.33+0.232026-08-02 → 2026-08-30
tmax00Znearest90d27 1 QC6.29 [5.14, 7.23]2.08-6.29 [-7.23, -5.14]6.74 [5.54, 7.50]0% [0%, 0%]4%11%-1.33+0.232026-08-02 → 2026-08-30
tmax00Znearest365d33 1 QC5.82 [4.55, 6.95]2.19-5.82 [-6.95, -4.55]6.33 [5.08, 7.29]0% [0%, 0%]6%18%-0.90+0.272026-01-02 → 2026-08-30
tmax00Znearestall33 1 QC5.82 [4.46, 6.93]2.19-5.82 [-6.93, -4.46]6.33 [4.92, 7.25]0% [0%, 0%]6%18%-0.90+0.272026-01-02 → 2026-08-30
tmax12Zbilinear30d27 1 QC6.71 [5.63, 7.56]2.15-6.71 [-7.56, -5.63]7.15 [6.20, 7.83]0% [0%, 0%]7%7%-1.48+0.212026-08-02 → 2026-08-30
tmax12Zbilinear90d27 1 QC6.71 [5.47, 7.74]2.15-6.71 [-7.74, -5.47]7.15 [5.91, 7.96]0% [0%, 0%]7%7%-1.48+0.212026-08-02 → 2026-08-30
tmax12Zbilinear365d32 1 QC6.01 [4.19, 7.39]2.50-6.01 [-7.39, -4.19]6.64 [5.01, 7.71]3% [0%, 12%]16%19%-1.09+0.122026-01-02 → 2026-08-30
tmax12Zbilinearall32 1 QC6.01 [4.15, 7.37]2.50-6.01 [-7.37, -4.15]6.64 [4.85, 7.69]3% [0%, 12%]16%19%-1.09+0.122026-01-02 → 2026-08-30
tmax12Znearest30d27 1 QC6.72 [5.60, 7.61]2.20-6.72 [-7.61, -5.60]7.18 [6.20, 7.88]0% [0%, 0%]7%15%-1.48+0.192026-08-02 → 2026-08-30
tmax12Znearest90d27 1 QC6.72 [5.39, 7.86]2.20-6.72 [-7.86, -5.39]7.18 [5.92, 8.04]0% [0%, 0%]7%15%-1.48+0.192026-08-02 → 2026-08-30
tmax12Znearest365d32 1 QC6.03 [4.19, 7.45]2.54-6.03 [-7.45, -4.19]6.67 [5.07, 7.77]3% [0%, 12%]12%25%-1.10+0.112026-01-02 → 2026-08-30
tmax12Znearestall32 1 QC6.03 [4.17, 7.42]2.54-6.03 [-7.42, -4.17]6.67 [4.85, 7.75]3% [0%, 12%]12%25%-1.10+0.112026-01-02 → 2026-08-30
tmin00Zbilinear30d27 1 QC1.64 [1.09, 2.23]1.24-1.10 [-1.66, -0.52]2.00 [1.30, 2.55]30% [15%, 43%]70%81%+0.56+0.662026-08-02 → 2026-08-30
tmin00Zbilinear90d27 1 QC1.64 [1.00, 2.40]1.24-1.10 [-1.82, -0.36]2.00 [1.19, 2.67]30% [12%, 47%]70%81%+0.56+0.662026-08-02 → 2026-08-30
tmin00Zbilinear365d33 1 QC1.87 [1.18, 2.66]1.47-1.32 [-2.11, -0.60]2.42 [1.48, 3.38]27% [12%, 41%]67%79%+0.52+0.622026-01-02 → 2026-08-30
tmin00Zbilinearall33 1 QC1.87 [1.19, 2.59]1.47-1.32 [-2.09, -0.59]2.42 [1.45, 3.32]27% [12%, 42%]67%79%+0.52+0.622026-01-02 → 2026-08-30
tmin00Znearest30d27 1 QC1.69 [1.10, 2.31]1.25-1.16 [-1.73, -0.55]2.07 [1.32, 2.61]30% [14%, 50%]74%81%+0.54+0.662026-08-02 → 2026-08-30
tmin00Znearest90d27 1 QC1.69 [1.01, 2.48]1.25-1.16 [-1.91, -0.40]2.07 [1.21, 2.75]30% [10%, 50%]74%81%+0.54+0.662026-08-02 → 2026-08-30
tmin00Znearest365d33 1 QC1.95 [1.23, 2.81]1.51-1.42 [-2.27, -0.67]2.54 [1.52, 3.58]27% [11%, 45%]70%76%+0.50+0.612026-01-02 → 2026-08-30
tmin00Znearestall33 1 QC1.95 [1.22, 2.72]1.51-1.42 [-2.22, -0.64]2.54 [1.51, 3.49]27% [12%, 47%]70%76%+0.50+0.612026-01-02 → 2026-08-30
tmin12Zbilinear30d27 1 QC2.15 [1.57, 2.67]1.35-1.86 [-2.34, -1.38]2.52 [1.88, 2.99]22% [10%, 36%]56%74%+0.42+0.632026-08-02 → 2026-08-30
tmin12Zbilinear90d27 1 QC2.15 [1.48, 2.73]1.35-1.86 [-2.42, -1.29]2.52 [1.76, 3.04]22% [8%, 38%]56%74%+0.42+0.632026-08-02 → 2026-08-30
tmin12Zbilinear365d32 1 QC2.23 [1.65, 2.73]1.55-1.60 [-2.28, -0.69]2.59 [1.97, 3.04]22% [10%, 36%]53%72%+0.43+0.592026-01-02 → 2026-08-30
tmin12Zbilinearall32 1 QC2.23 [1.64, 2.72]1.55-1.60 [-2.31, -0.78]2.59 [1.97, 3.03]22% [10%, 36%]53%72%+0.43+0.592026-01-02 → 2026-08-30
tmin12Znearest30d27 1 QC2.25 [1.65, 2.80]1.38-1.95 [-2.46, -1.45]2.63 [1.94, 3.16]22% [10%, 36%]52%74%+0.39+0.632026-08-02 → 2026-08-30
tmin12Znearest90d27 1 QC2.25 [1.57, 2.84]1.38-1.95 [-2.57, -1.37]2.63 [1.83, 3.20]22% [8%, 38%]52%74%+0.39+0.632026-08-02 → 2026-08-30
tmin12Znearest365d32 1 QC2.30 [1.71, 2.82]1.56-1.72 [-2.42, -0.84]2.68 [2.01, 3.19]22% [10%, 36%]53%72%+0.42+0.592026-01-02 → 2026-08-30
tmin12Znearestall32 1 QC2.30 [1.74, 2.81]1.56-1.72 [-2.45, -0.92]2.68 [2.04, 3.18]22% [10%, 36%]53%72%+0.42+0.592026-01-02 → 2026-08-30

+ model too warm · − model too cold · bias interval includes zero · ★ lowest MAE · = not distinguishable from the leader · ▼ significantly worse · n < 30 greyed and unranked · CI is a 95 % moving-block bootstrap interval.

Daily error, last 90 days

Daily forecast minus observed error in °F for GraphCast (IFS init) at KMSP, lead day 1, 00Z bilinear tmax+20+100-10-202026-08-02: -6.66 °F2026-08-03: -3.71 °F2026-08-04: -2.54 °F2026-08-05: -6.8 °F2026-08-06: -4.31 °F2026-08-07: -2.29 °F2026-08-08: -5.09 °F2026-08-09: -8.67 °F2026-08-10: -8.32 °F2026-08-11: -7.21 °F2026-08-12: -7.66 °F2026-08-13: -2.41 °F2026-08-14: -4.46 °F2026-08-15: -4.35 °F2026-08-16: -9.31 °F2026-08-17: -8.83 °F2026-08-18: -8.75 °F2026-08-19: -8.11 °F2026-08-20: -3.25 °F2026-08-21: -6.53 °F2026-08-22: -6.21 °F2026-08-23: -8.47 °F2026-08-24: -10.2 °F2026-08-25: -3.79 °F2026-08-26: -8.73 °F2026-08-29: -8.13 °F2026-08-30: -6.03 °F2026-08-022026-08-30°F, forecast − observed
Forecast − observed, °F, 00Z bilinear daily maximum. The dashed line is zero; the thick line is a centred 7-day mean, drawn only where the whole window is present. Gaps are days with no score, never interpolation. Hover a point for its date and value.
The same series as a table (29 days, newest first) · download CSV
Climatological dayError °F
2026-08-30-6.03
2026-08-29-8.13
2026-08-28—
2026-08-27—
2026-08-26-8.73
2026-08-25-3.79
2026-08-24-10.20
2026-08-23-8.47
2026-08-22-6.21
2026-08-21-6.53
2026-08-20-3.25
2026-08-19-8.11
2026-08-18-8.75
2026-08-17-8.83
2026-08-16-9.31
2026-08-15-4.35
2026-08-14-4.46
2026-08-13-2.41
2026-08-12-7.66
2026-08-11-7.21
2026-08-10-8.32
2026-08-09-8.67
2026-08-08-5.09
2026-08-07-2.29
2026-08-06-4.31
2026-08-05-6.80
2026-08-04-2.54
2026-08-03-3.71
2026-08-02-6.66

Error distribution

Distribution of the signed daily error in °F for GraphCast (IFS init) at KMSP, lead day 1-20 to -18 °F: 0 days-18 to -16 °F: 0 days-16 to -14 °F: 0 days-14 to -12 °F: 0 days-12 to -10 °F: 1 day-10 to -8 °F: 9 days-8 to -6 °F: 7 days-6 to -4 °F: 6 days-4 to -2 °F: 9 days-2 to 0 °F: 1 day0 to 2 °F: 0 days2 to 4 °F: 0 days4 to 6 °F: 0 days6 to 8 °F: 0 days8 to 10 °F: 0 days10 to 12 °F: 0 days12 to 14 °F: 0 days14 to 16 °F: 0 days16 to 18 °F: 0 days18 to 20 °F: 0 daysdashed: P50 |e| 6.03 · P90 |e| 8.74 °F-200+20 °Fn = 33 days
All scored days for 00Z bilinear daily maximum, 20 equal bins. The dashed marks are the 50th and 90th percentiles of the absolute error, mirrored on both sides. Hover a bar for its bin and count.
Bin counts as a table
From °FTo °FDays
-20.00-18.000
-18.00-16.000
-16.00-14.000
-14.00-12.000
-12.00-10.001
-10.00-8.009
-8.00-6.007
-6.00-4.006
-4.00-2.009
-2.00+0.001
+0.00+2.000
+2.00+4.000
+4.00+6.000
+6.00+8.000
+8.00+10.000
+10.00+12.000
+12.00+14.000
+14.00+16.000
+16.00+18.000
+18.00+20.000

Month by month

00Z bilinear daily maximum, newest first.

Monthn MAE °FBias °F
2026-08276.33 -6.33
2026-0163.54 -3.54

Availability

ModelScored period DaysCoverage
GraphCast (IFS init)2026-01-02 → 2026-08-30 33
GraphCast (IFS init)2026-01-02 → 2026-08-30 33
GraphCast (IFS init)2026-01-02 → 2026-08-30 32
GraphCast (IFS init)2026-01-02 → 2026-08-30 32

Compared with other models

Paired bootstrap on the days both models cover (90d, 00Z, bilinear, daily maximum). Negative means GraphCast (IFS init) has the lower MAE. "Distinguishable" means the 95 % interval excludes zero.

Other modelCommon days MAE difference °FDistinguishable
ECMWF AIFS Single 27 +2.70 [2.04, 3.22] yes
Aurora (GFS init) 1 -2.08 [-2.08, -2.08] yes
Aurora (IFS init) 2 +0.54 [-0.09, 1.16] no
FourCastNet v2 (GFS init) 1 +0.02 [0.02, 0.02] yes
FourCastNet v2 (IFS init) 2 +0.67 [-0.10, 1.44] no
NCEP GFS 27 +3.04 [1.97, 4.22] yes
GraphCast (GFS init) 1 -1.57 [-1.57, -1.57] yes
ECMWF IFS HRES 21 +3.10 [1.90, 4.07] yes
Pangu-Weather (GFS init) 1 +5.49 [5.49, 5.49] yes
Pangu-Weather (IFS init) 2 +2.52 [1.20, 3.83] yes
Persistence (baseline) 27 +3.62 [2.27, 5.06] yes

Cite this page

ShortCastCheck, KMSP graphcast_ifs lead 1, methodology v0.2, accessed 2026-08-31, https://castcheck.zifanzhang.com/station/KMSP/model/graphcast_ifs/lead/1/

LongCastCheck (2026). Station-level verification of raw weather-model 2 m temperature forecasts: KMSP (Minneapolis-St Paul), model graphcast_ifs, lead day 1. Methodology version 0.2, schema version 0.1, data through 2026-08-30. Accessed 2026-08-31. https://castcheck.zifanzhang.com/station/KMSP/model/graphcast_ifs/lead/1/

Accessed2026-08-31 (data through 2026-08-30; page built 2026-08-31T03:52:06+00:00)

JSON CSV all scores methodology v0.2 licence